TY - JOUR
T1 - Attention-based Convolutional Autoencoders for 3D-Variational Data Assimilation
AU - Mack, Julian
AU - Arcucci, Rossella
AU - Molina-Solana, Miguel
AU - GUO, Yi-Ke
N1 - Funding Information:
This work is supported by the EPSRC, UK Grand Challenge grant ?Managing Air for Green Inner Cities? (MAGIC) EP/N010221/1, by the EPSRC, UK Centre for Mathematics of Precision Healthcare EP/N0145291/1 and the EP/T003189/1 Health assessment across biological length scales for personal pollution exposure and its mitigation (INHALE). Thanks to Dr. Laetitia Mottet for the set up of the full model in Fluidity. M. Molina-Solana was supported by European Union's H2020 MSCA-IF (ga. No. 743623) and Athenea3i (ga. No. 754446) programmes.
Funding Information:
This work is supported by the EPSRC, UK Grand Challenge grant “Managing Air for Green Inner Cities” (MAGIC) EP/N010221/1 , by the EPSRC, UK Centre for Mathematics of Precision Healthcare EP/N0145291/1 and the EP/T003189/1 Health assessment across biological length scales for personal pollution exposure and its mitigation (INHALE). Thanks to Dr. Laetitia Mottet for the set up of the full model in Fluidity. M. Molina-Solana was supported by European Union’s H2020 MSCA-IF (ga. No. 743623 ) and Athenea3i (ga. No. 754446 ) programmes.
PY - 2020/12/1
Y1 - 2020/12/1
N2 - We propose a new ‘Bi-Reduced Space’ approach to solving 3D Variational Data Assimilation using Convolutional Autoencoders. We prove that our approach has the same solution as previous methods but has significantly lower computational complexity; in other words, we reduce the computational cost without affecting the data assimilation accuracy. We tested our proposal with data from a real-world application: a pollution model of a site in Elephant and Castle (London, UK) and found that we could (1) reduce the size of the background covariance matrix representation by O(103), and (2) increase our data assimilation accuracy with respect to existing reduced space methods.
AB - We propose a new ‘Bi-Reduced Space’ approach to solving 3D Variational Data Assimilation using Convolutional Autoencoders. We prove that our approach has the same solution as previous methods but has significantly lower computational complexity; in other words, we reduce the computational cost without affecting the data assimilation accuracy. We tested our proposal with data from a real-world application: a pollution model of a site in Elephant and Castle (London, UK) and found that we could (1) reduce the size of the background covariance matrix representation by O(103), and (2) increase our data assimilation accuracy with respect to existing reduced space methods.
KW - Attention networks
KW - Convolutional Autoencoders
KW - Variational Data Assimilation
UR - http://www.scopus.com/inward/record.url?scp=85089822637&partnerID=8YFLogxK
U2 - 10.1016/j.cma.2020.113291
DO - 10.1016/j.cma.2020.113291
M3 - Journal article
AN - SCOPUS:85089822637
SN - 0045-7825
VL - 372
JO - Computer Methods in Applied Mechanics and Engineering
JF - Computer Methods in Applied Mechanics and Engineering
M1 - 113291
ER -